REQUESTIONING OLD RUSSIAN AUTOBIOGRAPHICAL WRITINGS
REQUESTIONING OLD RUSSIAN AUTOBIOGRAPHICAL WRITINGS
REQUESTIONING OLD RUSSIAN AUTOBIOGRAPHICAL WRITINGS
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
The coefficients of the model verify our hypothesis that there is a positive connection<br />
between a population’s life expectancy and its level of prosperity. Differences in wealth were<br />
assessed on the basis of spending on indicative goods. The negative impact of unemployment on<br />
life expectancy also came as no surprise. A dummy variable was introduced to eliminate the<br />
distorting influence of the Tuva Republic, where average life expectancy is significantly lower.<br />
An important implication of this model is the constant negative impact of the<br />
consumption of pure alcohol on average life expectancy in the regions.<br />
We noted that alcohol abuse not only reduces life expectancy, but also has a significant<br />
impact on mortality from external causes. This indicator has a strong negative correlation with<br />
life expectancy, with a coefficient of -0.84. Regions with the least mortality from external causes<br />
are usually included in the list of regions with the highest life expectancy, and vice versa. At the<br />
same time, variations in mortality from external causes are significantly higher.<br />
In order to test the hypothesis that the level of alcohol consumption has a negative impact<br />
on health in both the male and female populations, we have built regression models where the<br />
dependent variables were indicators of life expectancy for men and women separately.<br />
Table 2.9.<br />
Result of regression analysis on (male) life expectancy (Russian regions)<br />
Dependent variable: Life expectancy (male), N=80<br />
Variable<br />
Coefficient<br />
Standard<br />
Deviation<br />
t-statistic<br />
Prob.<br />
Const 72.15743 1.005047 71.79509 0.0000<br />
Computers spending per capita 0.784776 0.375159 2.091851 0.0400<br />
Meat spending per capita -0.462243 0.138474 -3.338123 0.0013<br />
Sugar spending 5.06E-07 1.13E-07 4.462103 0.0000<br />
Unemployment rate 17.96891 6.266397 2.867503 0.0054<br />
Mortality rate from external reasons -0.036893 0.003813 -9.675733 0.0000<br />
Vodka consumption -0.147239 0.055514 -2.652285 0.0099<br />
Beer to vodka consumption ratio -0.170841 0.080180 -2.130706 0.0366<br />
Wine to vodka consumption ratio -2.725992 0.781676 -3.487367 0.0008<br />
Coefficient of determination R-squared 0.858288<br />
F-statistic 53.75181<br />
Prob(F-statistic) 0.000000<br />
Mortality from external causes, which allows us to separate the contribution of short-term<br />
alcohol effects on health and living standards from long-term (cumulative) effects, was<br />
introduced into this model as the independent variable. In this model, we also used data on<br />
consumption of alcoholic beverages by type to reflect the impact of the structure of alcohol<br />
consumption on average life expectancy. Converting these to ratios helped to avoid problems<br />
arising from multicollinearity in data on the consumption of different alcoholic beverages types.<br />
17